SOUTH COAST AIR QUALITY MANAGEMENT DISTRICT
METHOD 25.3
DETERMINATION OF LOW CONCENTRATIONNON-METHANE NON-ETHANE ORGANIC COMPOUND EMISSIONS
FROM CLEAN FUELED COMBUSTION SOURCES
MONITORING AND ENGINEERING BRANCHMONITORING AND ANALYSIS
March 2000
SCAQMD Method 25.3 March 2000
2
TABLE OF CONTENTS
List of Figures 3
1. Overview and Applicability1.1 Principle 41.2 Applicability 61.4 Limitations and Interferences 8
2. Sampling Apparatus and Field Equipment Preparation2.1 Sampling Apparatus 92.2 Sampling Reagents 122.3 Sampling Equipment Preparation 13
3. Field Procedures3.1 Pre-Test Determinations 183.2 On-site Equipment Assembly 183.3 Pre-test Leak Check 203.4 Sampling Operation 203.5 Reference Point Velocity 203.6 Post-test Procedures 22
4. Laboratory Procedures 4.1 Sample Receipt 254.2 Sample Purge 264.3 Apparatus and Reagents for TOC Analysis on Traps 274.4 Preparation of Standards and Reagents for TOC Analysis 284.5 TOC Analysis on Traps 294.6 TOC Analysis Quality Assurance (QA) Criteria 314.7 TOC Calculations 324.8 Apparatus and Reagents for TCA Analysis on Canisters 334.9 TCA Analysis on Canisters 374.10 TCA Analysis Quality Assurance (QA) Criteria 434.11 TCA Calculations 43
5. Engineering Calculations and Reporting5.1 Data Quality Checks 505.2 VOC Molecular Weight per Carbon Ratio 505.3 Bias Correction Factor 545.4 VOC Mass Emission Rate Calculation 55
SCAQMD Method 25.3 March 2000
3
LIST OF FIGURES
Figure 25.3-1 Preparation of Sampling Apparatus 16Figure 25.3-2 Condensate Trap Detail 17Figure 25.3-3 Sampling Apparatus During Sampling 23Figure 25.3-4 Field Data Sheet 24Figure 25.3-5 Inert Gas Purging 45Figure 25.3-6 Flow Diagram for TCA Analysis on Canisters 46Figure 25.3-7 Equipment Diagram for TCA Analysis 47Figure 25.3-8 Equipment Operation for TCA Analysis 48Figure 25.3-9 Example Chromatogram for TCA Analysis 49Figure 25.3-10 Molecular Weight per Carbon Ratios 56Figure 25.3-11 VOC Mass Emission Rate Calculation 57
SCAQMD Method 25.3 March 2000
4
METHOD 25.3
DETERMINATION OF LOW CONCENTRATIONNON-METHANE NON-ETHANE ORGANIC COMPOUND EMISSIONS
FROM CLEAN FUELED COMBUSTION SOURCES
Section 1 of 5
1. Overview and Applicability
1.1 Principle
The procedures used in this Volatile Organic Compound (VOC) source test method are
similar to the approach of Method 25.1, but have been modified for the purposes of
improving accuracy at low concentrations. The method eliminates positive interferences
for low concentration VOC due to high levels of stack carbon dioxide and moisture. As
with Method 25.1, duplicate gas samples are withdrawn from a source at a constant rate
through condensate traps (traps) followed by evacuated canisters. The method differs
from Method 25.1 in that stack condensate is collected at ice water (∼ 32 oF) temperature in
the traps as opposed to the lower dry ice temperature. For low concentrations, the ∼ 32 oF
traps have proven to be sufficient for trapping condensate and preventing unrecoverable
VOC from being collected in the canisters. With clean sources, since the condensate
consists largely of water, the traps consist of small impingers initially charged with ultra-
pure water as a heat transfer medium. Interfering carbon dioxide in the traps is purged
into the canisters using a ultra-pure grade inert gas. Particulate matter is prevented from
interfering with the method by using an in-stack filter. Since the water does, however,
have limitations on the amount of insoluble material that can be homogeneously retained,
the method is limited to VOC concentrations of less than 50 ppm as carbon (ppmC) or 25
ppmC in the trap section.
VOC concentration as Non-Methane Non-Ethane Organic Compounds (NMNEOC) is
determined by combining the results from the independent analyses of the condensate in
SCAQMD Method 25.3 March 2000
5
each trap and the gas in its associated canister. The traps are analyzed for total organic
carbon by liquid injection into an infra-red total organic carbon analyzer. The canisters
are analyzed for NMNEOC using the Method 25.1 approach. The analysis consists of
foreflush and backflush of a gas chromatography (GC) column followed by an oxidizer,
methanizer, and a flame ionization detector (FID). The GC separates the VOC component
from the sample; the oxidizer converts the VOC to carbon dioxide; the methanizer
converts the resulting carbon dioxide to methane. The results are determined by the FID
in units of parts per million by volume as carbon (ppmC). Carbon monoxide and fixed
gases (carbon dioxide and oxygen) can be determined by analysis of the canister portion of
the sample by SCAQMD Method 10.1.
The method is written to represent the configuration used during validation testing.
Mention of trade names in this source test method does not constitute endorsement by
SCAQMD; the model names and numbers are given as those used during the validation
phase of the method. Other manufacturers of equipment may be used subject to
demonstration of equivalency as approved by the SCAQMD.
A bias correction factor of 1.086 must be applied to the final results of this method. The
use of this bias correction factor is required by the USEPA as determined during the
validation phase of the method (refer to Section 5.3).
SCAQMD Method 25.3 March 2000
6
1.2 Applicability
This method replaces the method that was formerly known as SCAQMD Draft Method
25.2. Former Draft Method 25.2 has been removed from consideration due to inherent
shortcomings in its approach which have been proven to cause both a low bias and poor
precision. Source test results achieved by former Draft Method 25.2 are, therefore,
not considered as valid for SCAQMD purposes.
Method 25.3 measures low concentration VOC emissions as NMNEOC expressed as
ppmC. Since it is not adversely affected by the unpredictability of VOC composition in
combustion products, the method is particularly applicable to combustion processes. In its
total carbon approach, the method is not affected by compound specific instrument
response factor variables often encountered in other detection methods. This method is
applicable when the following conditions are met:
1. Combustion sources must use clean fuels. Clean fuels are defined as natural gas,
refinery fuel gas, butane, LPG, landfill gas, digester gas, methanol and ethanol.
2. The resulting concentration as measured by this method must be less than 50 ppmC or
alternatively 25 ppmC in the trap portion with a higher limit on the canister portion
evaluated case by case depending on the compounds present.
Supporting data has shown that for concentrations above 50 ppmC (or 25 ppm in the trap)
or for non-clean fueled combustion sources, a bias will occur due to limitations in the
condensate trap design (see Interferences). For these situations, refer to Method 25.1.
The method may be applied to sources of higher concentrations where exclusively water
soluble VOC is encountered. The applicability of the method to these situations must be
evaluated by the SCAQMD on a case by case basis.
SCAQMD Method 25.3 March 2000
7
The method may be used without the condensate trap and its associated procedures only
for ambient temperature sources where no combustion products or other sources of
moisture other than ambient are present. Additionally, the resulting concentration as
measured by this method must be less than 50 ppmC. The applicability of the method to
these situations must be evaluated by SCAQMD on a case by case basis.
The lower detection limit of the method is 1 ppm NMNEOC as carbon.
For determining mass emissions, a molecular weight per carbon ratio must be established
to account for bonded oxygen, hydrogen, chlorine, or other elements. Similarly for
converting ppmC to actual ppm, the average carbon number must be estimated. Section
five of this method provides guidelines for determining molecular weight per carbon.
This method assumes that methane and ethane are the only significant VOC exempt
compounds commonly found in combustion exhausts. The NMNEOC results can be
corrected for other exempt compounds when present, using an appropriate method
approved by SCAQMD, CARB, and EPA for determining exempt content.
SCAQMD Method 25.3 March 2000
8
1.4 Limitations and Interferences
In cases where combustion type control devices are used to control streams containing
exempt compounds as specified in SCAQMD Rule 102, a positive bias towards VOC will
occur if a correction is not made for the presence of exempt compounds.
Supporting data has shown a potential negative bias when sampling streams of over 50
ppmC due to design limitation of the condensate traps which were designed for lower
concentrations. This is believed to be caused by non-homogeneity in the water traps with
insoluble species present.
Sampling combustion sources burning non-clean fuels can cause a negative bias. This is
also due to limitations in the low-concentration trap design combined with the presence of
high molecular weight, semi-volatile, condensable, insoluble material which tends to
separate from the water in the trap.
Ammonia present in concentrations above 15 ppm can reduce the level of precision of this
method but does not cause a specific low or high bias.
Because of the low concentration of VOC encountered, contamination, if present can
cause a significant bias. The procedures of this method have been designed to eliminate
potential contamination. It is, however, imperative that the equipment cleaning and
sample handling procedures are carefully followed.
If samples are extracted from a stratified stream, a positive or negative bias may occur.
Samples must be extracted from well mixed locations or employ multi-point sampling (see
Field Procedures).
Procedures for minimizing the effects of any of the above interferences where applicable
are not all addressed in this method. Alternative procedures must be evaluated on a case
by case basis and are subject to SCAQMD approval.
SCAQMD Method 25.3 March 2000
9
METHOD 25.3
DETERMINATION OF LOW CONCENTRATIONNON-METHANE NON-ETHANE ORGANIC COMPOUND EMISSIONS
FROM CLEAN FUELED COMBUSTION SOURCES
Section 2 of 5
2. Sampling Apparatus and Field Equipment Preparation
2.1 Sampling Apparatus
The sampling system consists of an in-stack filter, a probe, a Teflon line, a condensate
trap, a flow controller, a vacuum gauge, a valve, and a canister (see Figure 25.3-1). The
sampling equipment can be constructed from commercially available components. The
internal volume of the entire sampling apparatus excluding the canister must not exceed
one percent of the canister volume in order to avoid dilution by the sample system dead
space. The following is a detailed description of the sampling system component
requirements.
a. In-Stack Filter
A ≤2 micron, 316 type stainless steel or other high temperature non-corrosive
material filter located at the stack end of the probe tip. The filter can be the small frit
type inserted into a 0.25 in. tube fitting connected to the probe tip.
b. Probe
Seamless stainless steel tubing, 0.25 in. outside diameter and cut to a length of half
the stack diameter or sufficient length to extend near the stack or duct center to avoid
dilution effects from the sampling port. When sampling, the probe is fixed with the
connector line end flush with the port entrance so that the stack gases heat the entire
probe length.
SCAQMD Method 25.3 March 2000
10
c. Condensate Trap
The condensate trap is designed as a small 4 ml glass bodied impinger. The body is a
commonly available 4 ml narrow screw top glass vial which is used not only as the
trap body but also for sample storage with its supplied Teflon lined screw cap. The
approximate dimensions of the vials are 1.8 in. total height, 0.6 in. o.d., and 0.3 in.
i.d. at the upper threaded opening. A 0.25 in. hole must be made in a spare screw cap
for affixing the condensate trap to the sampling assembly. Size 009 and a size 006
Viton “O” rings are used to seal and retain the glass vial to the sampling assembly.
See Figure 25.3-2 for specifications on condensate trap design. For sampling, the
condensate trap is charged with approximately 2 ml of hydrocarbon free water. The
4 ml trap size is sufficient for stack moistures of up to 25% by volume. For higher
stack moistures, the trap design must be scaled up accordingly.
d. Connector Line
Seamless Perfluoroalkoxy (also known as PFA, a type of Teflon) tubing, 0.125 in.
outside diameter x 0.026 in. wall thickness and cut to length of no more than 18 in.
The connector line is connected to the probe with a stainless steel tube reducer. The
other end of the connector line must extend into the condensate trap with a tapered
end having an opening of less than 0.020 in. so that small bubbles are formed in the
condensate trap. This tapered end can be formed by applying a point source of heat
to approximately one inch of a continuous section of tubing so that the temperature
of the section is heated to near the melting point. The opposite sides of the heated
section can be pulled apart while twisting torsionally to form a split at the heated
section. The narrow tips created at the ends of the tubing split can then be trimmed
appropriately to create the small opening. When assembled, the tubing extends from
the probe reducer to within 0.125 in. from the bottom of the condensate trap as
shown in Figure 25.3-2.
SCAQMD Method 25.3 March 2000
11
e. Ice water bath
A container is affixed to the sampling apparatus to hold ice and water used to cool
the condensate trap. The ice bath must be of sufficient cross section to surround the
condensate trap with an appropriate amount of ice to maintain the ice water
temperature (minimum 2.5 in. diameter). The ice bath must be approximately one
inch in height (for 4 ml trap size) so that the condensate trap vial connection remains
above the top of the ice bath container and the overflowing cold water will be below
the connector level. This is done to eliminate the risk of contamination. The ice bath
must also be positioned sufficiently high so that the water level of the bath is higher
than the water level inside the condensate trap.
f. Flow Rate Controller
A vacuum flow regulator, rotameter, fine orifice, or other flow regulator capable of
maintaining a constant flow rate (± 10%) at the probe tip over the sampling period.
The flow controller is located downstream of the condensate trap so that its function
is unaffected by the condensate. For flows regulated by rotameters or orifices with a
control valve, the control valve will require constant adjustment during sampling due
to the declining pressure differential. The control valve must be located between the
canister and the rotameter or orifice. For a critical orifice, where a control valve is
not used, sampling must be terminated if the vacuum in the canister drops below the
level where a constant flow cannot be achieved. This type of orifice can be prepared
using a GC syringe needle fixed concentrically into 1/4 in. stainless steel tubing.
Epoxy or silicone adhesive has been successfully used for this purpose. The desired
flow rate for one hour sampling time for a 6-liter canister is ~70 ml/min and can be
achieved using a short (appropriate) length of 0.0045” syringe needle. The flow rates
can be fine tuned by adjusting the length of the inner needle tube.
SCAQMD Method 25.3 March 2000
12
g. Vacuum Gauge
A stainless steel gauge cleaned for electronic use is specified for monitoring vacuum
in the tank and sampling system between the flow controller and sample flow valve
both during sampling and leak checks (0 to 30 in. Hg Vacuum).
h. Sample Flow Valve
Stainless steel bellows valve is used for starting, stopping, or regulating sample flow
and is located between the vacuum gauge and sample canister.
i. Sample Canister
The electro-polished stainless steel canister has a volume of 6 ± 0.5 liters. The
canister volume is determined to the nearest 10 ml as described in section 3.8.
i. Sampling Assembly
The assembled sampling apparatus in its “ready for transport” configuration is shown
in Figure 25.3-1. An exploded view of the sample line and condensate trap assembly
is shown in Figure 25.3-2. A stainless steel quick connector is useful for connecting
and disconnecting the canister from the remainder of the sampling assembly.
2.2 Sampling Reagents
The condensate trap is initially charged with approximately 2 ml hydrocarbon free water
such as deionized or distilled water. This water must have a TOC content of less than one
ppmC.
SCAQMD Method 25.3 March 2000
13
2.3 Sampling Equipment Preparation
2.3.1 Sampling Equipment Cleaning
The sampling equipment preparation must be performed in a clean indoor
laboratory type environment and not in the field. All equipment that contacts the
sample excluding the canister, but including the remaining equipment listed in
section 2.1 and other equipment that contacts the sample such as connectors and
end caps, must be thoroughly cleaned as follows: Soak the equipment in non-
residue, rinsable type laboratory glassware detergent and water. Scrub all
accessible surfaces and remove all visible surface residues. Rinse the equipment
thoroughly first with tap water then with deionized water. At this point onward, be
certain not to touch any part of the internal sample path or open connection ends
with any object that has not been cleaned using the above procedure and
particularly not at anytime with hands or fingers. Use powder-free latex gloves
while handling cleaned equipment. Blow the equipment pieces dry with ultra-pure
grade air (< 0.5 ppm hydrocarbon) while holding the pieces by the outside surface
which does not contact the sample. Under no circumstances should uncleaned
compressed air be used for drying parts due to the possibility of entrained
compressor lube oil or other droplets causing contamination. After drying, the
equipment excluding the canister can be assembled as in Figure 25.3-1 using a
clean empty glass vial on the condensate trap assembly during transportation to the
field. Clean the probe and filter assembly by exposing to elevated temperatures
using an open flame while passing air through the assembly. Gradually move the
flame along the entire length of the probe at a rate so that the probe is heated to a
glowing orange in each section contacted by the flame, then allow to cool. Seal the
open ends of the probe and sampling assembly using a clean cap or foil.
SCAQMD Method 25.3 March 2000
14
2.3.2 Canister Preparation
The following equipment are needed for canister preparation:
a. Manometer
Must be capable of measuring pressure to within 1 mm Hg in the 0-900 mm
range. Manometer must be NIST traceable.
b. Vacuum Pump
Capable of producing a full 30 in. Hg of gauge vacuum (50 ppmC were encountered. Clean the Summa polished
canisters by sequential filling with pure nitrogen gas and evacuating to
approximately 3 mm Hg. It has been determined that ten cycles are sufficient to
result in a
SCAQMD Method 25.3 March 2000
15
The sample tank is leak checked by isolating it from the vacuum pump and
allowing the tank to sit for at least 10 minutes. The tank is acceptable if no change
in tank vacuum is noted.
2.3.3 Condensate Trap Vial Preparation
Store 200 to 300 ml of the hydrocarbon free water in a clean glass jar with a glass
stopper at normal refrigerator temperature (approximately 5 oC). Analyze the water
for TOC content at ice water temperature as in Section 4. If the TOC content is
less than 1 ppmC, keep the water for future use. If the TOC content is more than 1
ppmC, replace the water, and repeat the process. Label both a clean threaded 4 ml
glass vial and a threaded cap as a set. Tare and record the vial and cap set. Fill the
vial with ∼ 2 ml of the hydrocarbon free water (approximately half full) and tightly
replace the corresponding cap. Prepare an adequate number of vials for the
deployment of the duplicate samples that will be collected. Extra vials must be
prepared for reagent blanks and connector line rinses. It is important that plenty of
water from the same batch is left in the stoppered glass jar for use during analysis.
The vials and the remaining water in the stoppered glass jar are then stored at
refrigerator temperature in the laboratory until transport to the field.
SCAQMD Method 25.3 March 2000
16
Figure 25.3-1 Preparation of Sampling Apparatus
SCAQMD Method 25.3 March 2000
17
Figure 25.3-2 Condensate Trap Detail
SCAQMD Method 25.3 March 2000
18
METHOD 25.3
DETERMINATION OF LOW CONCENTRATIONNON-METHANE NON-ETHANE ORGANIC COMPOUND EMISSIONS
FROM CLEAN FUELED COMBUSTION SOURCES
Section 3 of 5
3. Field Procedures
Individual sampling runs shall consist of duplicate simultaneous samples as described in this
section. The descriptions are provided for individual samples in the duplicate set for purposes of
simplicity. Condensate trap blanks are required for use during analysis. By following the
equipment cleaning and canister preparation procedures, full field blanks are not required but
may be employed if deemed necessary. Field blanks would consist of a full sampling assembly
handled and analyzed identically to the actual samples with the exception that samples are not
drawn into the containers. Results from sampling must not be corrected using either field blanks
or any type of ambient sampling for reporting purposes. The results from either field blanks or
ambient samples may be reported along with the sampling results.
3.1 Pre-test Determinations
Samples must be taken at well mixed and uniform locations, i.e. far from situations
causing stratification such as duct junctions, addition of dilution air, combustion zones, or
other flow disturbances that may alter the concentration profile. Alternatively, an
approved stratification check (refer to SCAQMD Source Test Manual Chapter X) using a
portable hydrocarbon analyzer may be used to check for stratification of less than 10% or
for a representative sampling point within a stratified duct. Multi-point sampling can
alternatively be employed but will require multiple concurrent samples with associated
probe lengths due to in-stack probe heating requirements.
SCAQMD Method 25.3 March 2000
19
The required sampling time interval is dependent on the applicable rule or permit
condition that is to be evaluated. In most cases when not specified, a full one hour
sampling period will be required since results will be used to determine emissions in lb/hr.
Sampling should begin and end only when the process has been operating for a sufficient
length of time and steady state operation can be assured. A steady state is defined as
operating at constant operating temperature, feed rate, fuel rate, product application or
throughput rate, etc., and that the production rate is steady throughout the process. For
batch or cyclic processes, the sampling period must encompass at least one complete cycle
or batch. The sampling period must also begin and end at the same point in the cycle so
that portions of the cycle are not over or under-represented.
3.2 On-site Equipment Assembly
Once in place at the sampling location, the equipment can be assembled as shown in
Figure 25.3-3 with the probe connected but not inserted into the stack. Care must be taken
during this step to avoid contamination of the internal surfaces of the condensate trap parts
by any contact with objects or dust in the area. The condensate trap water vials must be
chilled for a minimum of 5 minutes before sampling. The chilled condensate trap vial is
attached by removing the empty vial placed on the assembly during transport and
replacing with the water filled sample vial. The empty vial is then capped with the water
vial lid so that the combination is kept clean while not in use during sampling. Once the
vial is attached to the condensate trap assembly, the equipment must remain in the upright
position so that the condensate trap water does not drain out of the condensate trap
assembly into the flow controller.
The condensate trap can then be positioned with ice in the ice bath The position of the ice
bath relative to the condensate trap is such that the vial connection will be above the top of
the ice bath container so that the overflowing ice and cold water will be below the
connector level. The ice bath must also be positioned sufficiently high so that the water
SCAQMD Method 25.3 March 2000
20
level of the bath is higher than the water level inside the condensate trap. After
completing the assembly, record the vial and canister identification numbers on the field
data sheet as in Figure 25.3-4.
3.3 Pretest Leak Check
A pretest leak check is required. After assembling the sampling system as shown in
Figure 25.3-3, make certain that the fitting at the probe that holds the in-stack filter is
tightly capped. The leak check is performed by opening and closing of the sample flow
valve so that the valve is partially open for a sufficient amount of time to introduce the
canister vacuum to the remainder of the system. Immediately after the sample flow valve
is closed, the vacuum gauge may initially drop numerically in vacuum if a restricting
orifice is used as a flow controller. The vacuum drop should cease at numerically above
10 in. Hg. At this point a cease in movement of the vacuum gauge for a period of ten
minutes indicates an acceptable leak check on the sampling system. Additionally, when
sampling is initiated, the vacuum gauge must indicate a canister vacuum of numerically
greater than 28 in. Hg. If this initial vacuum is numerically less than 28 in. Hg, a leak in
the canister subsequent to its evacuation is indicated.
3.4 Sampling Operation
Uncap the filter fitting at the probe tip and place the probe in the stack with the opening of
the probe tip tangent to the stack flow. Clamp or fasten the probe in place so that the
entire stainless steel probe is within the heat of the stack and as far into the stack as
possible while avoiding melting the PFA connector line. The purpose of this probe
placement is to ensure that no condensation occurs in the probe. Condensation in the PFA
connector line is, however, acceptable. If present, the condensation should begin to form
after the junction of the probe to the PFA connector line. This can be verified by visual
observation of the condensation through the semitransparent PFA material. Clean the port
SCAQMD Method 25.3 March 2000
21
as much as possible before inserting the probe. When inserting the probe into the stack,
care must be taken so that the probe opening does not contact the stack port internal
surface residues or residues on the internal stack wall. Seal the port around the probe so
that ambient air does not dilute the stack gases.
If the process that is being sampled is operating under normal representative operating
conditions or the conditions specified by the permit conditions, sampling may begin. To
begin sampling, open the sample flow control valve and maintain a steady flow that varies
by no more than 10% so that the canister is filled from its full 30 in. Hg vacuum to a
numerical vacuum of 2 - 15 in. Hg over the specified sampling period as determined in
section 2.3.
Immediately after sampling has commenced, record the initial canister vacuum and clock
time. If this initial vacuum is numerically less than 28 in. Hg, then the sample is
invalidated. Provide ice in the ice bath during sampling to maintain a constant ice bath
temperature of ∼ 32 oF. Record the canister vacuums at regular intervals (15 minute
recommended) during sampling as an indicator of constant flow into the canisters. Fill in
the remaining information as prompted by the field data sheet in Figure 25.3-4. At the end
of the sampling period, record the final vacuum and clock time, then close the sample flow
valve. Remove the probe from the stack, note the condition of the in-stack filter, and
recap the probe at the filter fitting.
If the sampling is interrupted due to a shutdown in the process being sampled or for an
upset of normal or specified operation, close the sample flow valve to interrupt sampling.
Record vacuum gauge readings and clock time. When the source resumes the normal or
specified operating conditions, sampling may resume by reopening the sample flow
control valve.
SCAQMD Method 25.3 March 2000
22
3.5 Reference Point Velocity
If a flow rate is to be measured for determining mass emissions, monitor velocity at a
reference point during sampling. Take velocity readings at five minute intervals during
the sampling period, or more often when the velocity or temperature fluctuates by more
than 20 percent. Use the ratio of average reference point readings during sampling to
average reference point readings during the traverse to correct the average stack velocity
during the traverse. This is done so that average concentration measured during sampling
corresponds to the average flow rate experienced during sampling.
3.6 Post Test Procedures
Immediately after sampling, perform a post test leak check as in section 2.6 with the
maximum vacuum that can be achieved with the vacuum remaining in the canister.
After the post-test leak check, disconnect the PFA line from the probe. Rinse the
condensate present in the line into the condensate trap with 0.5 to 1.0 ml of hydrocarbon
free water. This is accomplished by introducing a small amount of the remaining tank
vacuum to the line while dipping the open end of the line briefly into a spare vial of the
hydrocarbon-free water. During this step, observe the water level in the trap and avoid
over-filling the trap to avoid the water being drawn into the flow controller. After the
connector line has been flushed, the condensate trap body is disconnected, capped, sealed,
and stored at approximately 32 oF. Alternatively the connector line can be capped and the
condensate trap left connected to the sampling assembly. If the condensate trap is left
connected, it does not need to be stored at 32 oF but must not be allowed to exceed 85 oF.
SCAQMD Method 25.3 March 2000
23
����������������
Figure 25.3-3 Sampling Apparatus During Sampling
SCAQMD Method 25.3 March 2000
24
Test No. Date
Company Name Recorded by
Sampling Location
METHOD 25.3 FIELD DATA SHEET
Pre-Test Leak Check: Post-Test leak CheckGauge Vacuum / in. Hg Gauge Vacuum: / in. HgLoss in 60 seconds / in. Hg Loss in 60 seconds: / in. Hg
Reference Point #
Sample Data Reference Point Data
Sample#1
Sample#2
Time VelocityHead
(in.H20)
Temperature(oF)
Canister No.
Trap No.
Controller No.
Location within Stack
Initial Time
Initial Vacuum (in. Hg)
Intermediate Time
Intermediate Vacuum (in. Hg)
Intermediate Time
Intermediate Vacuum (in. Hg)
Intermediate Time
Intermediate Vacuum (in. Hg)
Final Time
Final Vacuum (in. Hg)
Observations
Figure 25.3-4 Field Data Sheet
SCAQMD Method 25.3 March 2000
25
METHOD 25.3
DETERMINATION OF LOW CONCENTRATIONNON-METHANE NON-ETHANE ORGANIC COMPOUND EMISSIONS
FROM CLEAN FUELED COMBUSTION SOURCES
Section 4 of 5
4. Laboratory Procedures
The analyst must demonstrate prior to initial use that each of the analyzers used in this method is
capable of measuring low concentration NMNEOC. For the canister analysis this includes a
demonstration of proper separation, oxidation, reduction, and measurement. This demonstration
must also prove that the equipment can resolve lower concentration standards at just above the
lower detection limit (1 ppm) of this method. Achieving low concentration analysis for
NMNEOC by this method requires contaminant free equipment, an appropriate baseline
subtraction, and an appropriate range of calibration. This demonstration of the analyzers’
performance must be approved by the SCAQMD laboratory for use in this method.
4.1 Sample Receipt
a) Check the correctness of labels, number of samples vials, number of canisters, and
“Chain of Custody” forms for completeness of information.
a) Inspect the water sample vials for leakage. The canister gauge reading (if equipped)
should be 2 -15 in. .Hg, otherwise make a note.
a) Sample delivery personnel sign and date to relinquish the samples.
a) Laboratory personnel sign and date to receive the sample.
a) Store the sample vials in a clean refrigerator, and the canisters in a secured area. The
vials must be purged within 24 hours from sampling. The analysis must be performed
within ten days from sampling.
SCAQMD Method 25.3 March 2000
26
4.2 Sample Purge
a) Allow the sampling canisters to equilibrate to room temperature; then, using the
calibrated high precision manometer specified in Section 2.3.2(a), measure the
pressure of each canister to the nearest 1 mm Hg. The pressure should be similar to
the final sampling pressure as indicated by the sampling gauge. If a significant
pressure loss is observed indicating a leak, invalidate the sample. Invalidate the
sample when the absolute return pressure is less than 200 mm Hg. Record this
pressure as return pressure (Pr) before proceeding to the purge step.
b) Reassemble the sample vial to the remainder of the sampling assembly. If the
condensate trap vial was left connected to the sampling apparatus, it must be quickly
capped when disconnected to check canister vacuum, and kept upright until
reconnection after the received pressure has been taken. When the sampling apparatus
is reassembled, a leak check must be performed as in Section 3.3.
c) Connect the probe tip to a source of ultra pure grade nitrogen or argon and introduce a
flow of slightly greater than that of the sampling rate at the ultra pure gas source. The
gas source must contain a tee that is open to the atmosphere such that excess pressure
is bled to the atmosphere. Refer to Figure 25.3-5 for a schematic of this configuration.
d) Open the sampling canister valve and allow the pure gas to purge through the
assembly and into the canister. The minimum purging gas flow rate is 25 to 30
ml/minute. The bubbling characteristics should be similar to that encountered during
sampling. Allow the gas to purge for 10 minutes or until the vacuum drops to 2 in. Hg
numerically lower than the received vacuum.
SCAQMD Method 25.3 March 2000
27
e) After the purge period, close the purge gas valve first and allow any residual pressure
to vent through the purge gas line tee before closing the canister valve to avoid back
flushing the condensate trap assembly.
f) Remove the sample vial and cap securely. The glass vial is then analyzed for TOC or
stored at refrigerator temperature until analysis.
g) Remove the sampling assembly then pressurize the canister with pure argon or
nitrogen gas to a pressure between 860-910 mm Hg. Shut off pressure from pure gas
source, wait until reading is stable, record the reading as final pressure (Pf).
h) Disconnect the canister from the purge gas line and seal. The sample canister is
analyzed by TCA or stored until analysis.
4.3 Apparatus and Reagents for TOC Analysis on the Traps
4.3.1 Shimadzu TOC-5000 Analyzer
The TOC-5000 analyzer is automated. Total organic carbon (TOC) is measured
by the difference between total carbon (TC) and inorganic carbon (IC). TC,
containing both organic and inorganic carbon, is measured by oxidizing an
aliquot of sample with a Platinum catalyst at 650 ± 5 oC using an air carrier/
oxidizer. The CO2 gas is quantified against the stored calibration curve by a
non-dispersive infra-red (NDIR) detector. IC is measured by injecting an aliquot
of sample into a phosphoric acid (H3PO4) vessel. The CO2 released from
acidification of inorganic carbonaceous compounds with the acid is sparged with
air and then quantified the same way as CO2 from TC. The difference of the two
results is TOC.\
SCAQMD Method 25.3 March 2000
28
4.3.2 Other Apparatus and Reagents for TOC Analysis
250 ul glass syringe
Glass vials, 4 and 15 ml size with Teflon lined screw caps
Refrigerator set to a temperature of approximately 5 oC
Ice water bath
Analytical balance capable of weighing to 0.1 mg
Laboratory glassware as need
Deionized (DI) water containing
SCAQMD Method 25.3 March 2000
29
For a 1000 ppmC TOC stock standard, mix 0.2125 g KHP with a balance of DI water
and make 100g of solution. Cap tightly, and store in a refrigerator. Discard after two
months.
For a 1000 ppmC TC stock standard, mix 0.3497 g NaHCO3 and 04412 g of Na2CO3,
with a balance of DI water to make 100 g of solution. Cap tightly, and store in a
refrigerator. Discard the solution if a fibrous or flaky material appears.
Prepare working standards for TOC from the stock standard solutions; accurately weigh
aliquots of stock standard in a tarred 15 ml size, screw cap vial. Add DI water to about
80% capacity, reweigh the total solution. Calculate the concentration of working
standard using Equation 2:
ppmC of working standard = ppmC stock standard x Ws/Wts (2)
where Ws = weight in grams of stock standard
Wts = weight in grams of total solution
The recommended concentrations for TOC working standards are 10 ppmC, 40 ppmC
and 100 ppmC. TC standards are prepared in the same manner.
A TOC mixture for QC (QC standard) is prepared in the same manner as the standards
choosing organic compounds in the same class that represent the expected compounds
in the sample. The content of the stock solution may not be given in units of ppmC for
an individual component. For example, formaldehyde in water contains 37%
formaldehyde. Add concentrations of individual components to get the total
concentration as carbon.
SCAQMD Method 25.3 March 2000
30
4.5 TOC Analysis of Traps
a) Take the sample vials including field blank from the refrigerator one at a time, wipe
off any water on the vial prior to opening.
b) If less than 4 ml of water is present in the vials, open each vial add DI water (which
had been used for pre-field sampling preparation) to ~4 ml, cap the vial, and allow it
to equilibrate to room temperature.
c) Dry each sample vial, and weigh. Return all vials to the refrigerator.
d) The TOC analyzer is calibrated, prepared for sample analysis, and run according to
Manufacturer’s Instruction Manual. The following steps are applied generically the
TOC analysis.
e) Prepare an adequate ice-water bath and replenish ice as required during the entire
analysis.
f) Each analytical run for all DI water blanks, standards, field blanks, samples, and
controls consists of three injections/run with three wash cycles. These are all
performed at ice-water bath conditions. Perform the first run using the laboratory DI
water as a cleanup and a lab blank. Run the DI water until the average TOC of three
injections is less than 1 ppmC by using the previously stored calibration files.
g) Run TOC and TC standards and store in the calibration files according to the
calibration instruction in the manual. Multiple calibration files can be stored for the
various levels anticipated. Experience has shown that it is not required to run new
standards and new calibration on every batch of samples.
SCAQMD Method 25.3 March 2000
31
h) For each batch, initially check the validity of stored calibrations by running a TOC
and a TC standard at expected concentrations. Rerun the standard solutions if the
average area count of a standard is greater than ±2% of the calibration.
i) The analysis for each batch of samples is run in the following order:
Run a QC standard in the range of expected concentration.
Run the field blank.
Run the samples.
Run QC standards that bracket the sample concentrations.
4.6 TOC Analysis Quality Assurance (QA) Criteria
a) The precision of the TOC analysis must be 10% or less as determined using the
percent coefficient of variation (COV) from the three injections calculated as follows:
COV = 100 x (standard deviation / mean)
Where: The standard deviation of TOC is determined from the square root of the sum of the
squares of the standard deviations for TC and IC.
b) Accuracy of the QC sample as % difference from the prepared concentration must be
within ±10%.
c) TOC of the field blank concentration can be reported along with the results but not
used to correct the results. The field blank is typically equal or higher than lab DI
water blank (typically ~ 1 ppmC).
d) IC concentration of sample should be less than 10 ppmC (typically ~2 to 5 ppmC),
otherwise, make a note on the report.
SCAQMD Method 25.3 March 2000
32
4.7 TOC Calculations
Subtract the laboratory DI water blank TOC (not field blank) from the average of the three
sample analyses to yield a result for TOC in ug/ml of condensate trap water (Ci).
Calculate the amount of organic carbon as part per million by volume (ppmv) as gaseous
carbon in the sample using the following equation:
Cw = (Ci x Vi x Pa x Vid)/(Vc x Pr x Ac)
where:
Ac = Atomic weight of carbon (12.01 g/mol)
Cw = gaseous concentration of TOC as ppmv in condensate trap water
Ci = TOC concentration in ug/ml of condensate trap water
(Assume TOC concentration ug/g = ug/ml at 4oC)
Vi = volume of condensate trap water in ml
Vid = volume of ideal gas per mole at 25oC (24.4652 liters/mole)
Vc = volume of the SUMMA canister in liters
Pa = atmospheric pressure in mm Hg (760 mm Hg)
Pr = return pressure in mm Hg
SCAQMD Method 25.3 March 2000
33
4.8 Apparatus and Reagents for TCA Analysis by GC/FID on Canisters
4.8.1 TCA System
The Total Combustion Analysis (TCA) system consists of gas chromatography
(GC) modified with a backflush valve with reversed flow capability for back-
flushing the trapped NMNEOC. It is also equipped with a catalytic oxidizer, a
catalytic reducer, a flame ionization detector (FID), and a data handling system.
A gas sample is injected, using a 1 ml fixed loop 6-port gas injection valve, onto
a dual packed column. The NMNEOC analyzer is a semi-continuous GC/FID
analyzer capable of: (1) separating carbon monoxide (CO), carbon dioxide
(CO2), methane (CH4), ethane (C2H6), ethylene (C2H4), and NMNEOC, (2)
oxidizing the NMNEOC to CO2, and CO to CO2, (3) reducing the resulting
CO2 to CH4, (4) quantifying the CH4, and (5) after ethane elutes from the GC
column, the column is heated and backflushed to release remaining organic
compounds. The resulting CH4 is quantified against a stored standard curve by
the FID detector. See Figure 25.3-6 for a flow chart of the instrument. The
instrumentation system flow diagram is shown in Figure 25.3-7.
The analytical equipment are available commercially or can be constructed from
available components by a qualified instrument laboratory. The analyzer
consists of the following major components:
a. Sample Injection System
A heated six-port valve injector fitted with a 1 ml sample loop is
recommended. The sample loop consists of a sufficient length of 1/16 in.
stainless steel tubing so that the desired internal volume is achieved. The
installation of the valve is depicted in Figure 25.3-7. The sample
injection port and sample loop must be equipped with a heating
SCAQMD Method 25.3 March 2000
34
mechanism that maintains the specified temperature of 150 ± 5 oC. The
six port valve is located in the column oven.
b. Separation Column(s)
The gas chromatographic system consists of a two part column capable
of separating carbon monoxide (CO), carbon dioxide (CO2), methane
(CH4), ethane (C2H6), ethylene (C2H4), and NMNEOC. The two part
column consists of Tenax GC 80/100 mesh in a 1 ft. length of 1/8 in.
stainless steel tubing, in series with Chromosorb 106 80/100 mesh in a 6
ft length of 1/8 in. stainless steel tubing. The NMNEOC is trapped in the
Tenax section of the column while the remaining compounds are eluted
through the Chromosorb section. The backflush procedure is also
performed through both sections of the column. The column is contained
in an oven capable of performing the temperature ramping as specified in
Section 4.9. A heated four port valve is used to control flow direction
through the column as shown in Figure 25.3-7.
c. Oxidation Catalyst
A catalyst system capable of oxidizing CH4 to CO2 with at least 95
percent efficiency is acceptable. The oxidation catalyst system consists
of 15% chromium III oxide on 4 mm alumina pellets packed in the center
4 in. section of a 12 in. length of 1/4 in. diameter Inconnel tubing. The
remaining space on both ends of the tubing is packed with quartz wool
for retention of the catalyst. The oxidation catalyst is contained within a
heating device capable of maintaining a temperature of 650 ± 5 oC. A
four port valve is used to control flow during either oxidation or
SCAQMD Method 25.3 March 2000
35
regeneration modes of the oxidation catalyst. The installation of the
valve is depicted in Figure 25.3-7.
d. Reduction Catalyst (Methanizer)
A catalyst system capable of reducing CO2 to CH4 with at least 95
percent efficiency is acceptable. The reducing catalyst consists of nickel
on Gas Chrom R 80/100 mesh. To prepare the material, first dry the Gas
Chrom R 80/100 at 120 oC overnight. Allow to cool to room
temperature in a dessicator. Dissolve 1 g of nickelous nitrate in 30 ml
deionized water. Slowly add 10 g of the dried Gas Chrom R 80/100
mesh with constant stirring. Heat to dryness on a hot plate, then dry
overnight at 230 ± 5 oC. Allow again to cool to room temperature in a
dessicator. The catalyst is then packed in the center 4 in. section of a 12
in. length of 3/16 in. diameter Inconnel tubing. Each of the remaining
space at either end of the tubing is packed with quartz wool. The
reduction catalyst is contained within a heating device capable of
maintaining a temperature of 380 ± 5 oC. Reduction gas (hydrogen) is
supplied to the nickel catalyst tube by a tee fitting between the oxidation
and nickel catalyst tubes at a flow rate of approximately 45% of the total
final flow.
e. Flame-Out Buffer
The flame out buffer consists of Haysep Q 80/100 mesh packed in a 6 ft.
length of 1/8 in. diameter stainless steel tubing. The flame out buffer is
maintained at the FID detector temperature of 220 ± 5 oC..
SCAQMD Method 25.3 March 2000
36
f. FID
An FID with a linear response (+ 5 percent) over the operating range of
0.5 to 50 ppm CH4 is acceptable.
g. Data Recording System
Digital integration system compatible with the FID is used for
permanently recording the analytical results. The system must have a
software program capable of point to point baseline subtraction from the
standard and sample runs.
h. Sample flow valves
Three multi-port valves are needed to accomplish the sample, carrier, and
purge gas flow paths in this method. As specified in sections 4.8.1a,
4.8.1b, and 4.8.1c, a six port valve is used in the sample injection system,
a four port valve is used to control flow through the separation columns,
and a four port valve is needed for regeneration of the oxidation catalyst.
Figure 25.3-7 depicts the configuration of these four valves and the
position during each of two modes for each valve.
i. Syringes
Gas tight syringes, 30 ml and 100 ml capacities.
j. Reagents
1. Chromatographic grade helium as carrier gas.
2. Reagent grade hydrogen for reduction of CO2 and FID fuel.
3. USP breathing grade hydrogen-free air for FID combustion.
4. Three point NIST traceable CO calibration standards
SCAQMD Method 25.3 March 2000
37
5. Three point NIST traceable CO2 calibration standards
6. NIST traceable 1 ppm, 3 ppm, 10 ppm, and 100 ppm CH4 standards
in pure nitrogen
7. NIST traceable 1 ppm, 3 ppm, 10 ppm, and 100 ppm C2H4 standards
in pure nitrogen
8. NIST traceable 1 ppm, 3 ppm, 10 ppm, and 100 ppm iso-butane
standards for backflush in pure nitrogen
9. A high purity CO2 (approximately 12% to 15% in pure hydrogen free
nitrogen) is required as a background determination for high CO2
sample.
10. Other standards as required
Note: Alternatively multi-component gases can be used for each
concentration.
4.8.2 Other Apparatus and Reagents for TCA Analysis
Optionally the entire TCA system may be automated to control the temperatures,
valving, and detector attenuation using a computer and control software. This
system may feature analog/digital interface and a computer or an integrator for
the data handling system.
4.9 TCA Analysis on the Canisters
4.9.1 Instrument Parameters and Gas Flow Rates
Set instrument parameters as follows:
Sample Injection Port/Loop : 150 ± 5 oC
Detector : 220 ± 5 oC
Oxidation catalyst : 650 ± 5 oC
SCAQMD Method 25.3 March 2000
38
Reduction catalyst : 380 ± 5 oC
Heated transfer lines : 105 ± 5 oC
Sample flow valves : inside separation column oven
Separation Column Oven
Initial : 50 ± 2 oC for 8 minutes starting at injection
Ramp : Increase at a rate of 50 oC/min for 2 minutes
Final : 150 ± 5 oC for 5 minutes
Column Bake-Out : 190 ± 5 oC for 2 minutes
Gas flow rates:
Helium Carrier : 30 ml/min
Oxidation Catalyst Regeneration Air : 100 ml/min
Oxidation Catalyst Air : 180 ml/min
Methanizer Hydrogen : 12.3 ml/min
FID Hydrogen : 30 ml/min
FID Air : 300 ml/min
4.9.2 Equipment Conditioning
a) Establish all initial temperatures and gas flow rates as specified above.
b) Switch the valving so that the GC carrier stream is routed through the
oxidation catalyst system (Valves 1 and 2 of Figure 25.5-7 in Position 1) for
5 minutes.
c) Switch the valving so that the GC carrier stream is routed through both the
oxidation catalyst and the reduction catalyst system (Valve 3 of Figure 25.5-
7 in Position 1).
d) Turn on the detector air and hydrogen gases, then ignite the detector. The
detector attenuation is set at 8 and the range at 12.
e) Flush the 30 ml size sample syringe five times with ultra pure nitrogen.
SCAQMD Method 25.3 March 2000
39
f) Flush ultra pure nitrogen gas through the sampling connector fitting for 30
seconds for cleaning purposes.
g) Clean the injection system by flushing at least three syringe volumes of ultra
pure nitrogen gas.
4.9.3 TCA Procedure
Reproduce exactly the timing of valve switching and column temperature
changes for each blank, sample, and standard runs in a series. The following is
the sequence of events that occur during a single injection of standard, baseline,
or sample; the required sequence in which standards blanks and samples are
injected is given in Section 4.9.4. Refer to Figure 25.3-7 for references to the
valve and position numbers as indicated in parentheses and to Figure 25.3-8 for
a summary of the equipment operation.
a) Verify that the column temperature is 50 ± 2 oC and that the sample loop
valve is switched so that the carrier gas is routed through the column (valve
1 in position 1). Also verify that the column valve is switched so that the
carrier flows in the forward direction (valve 2 on position 1) and that the
oxidizer valve is switched so that the carrier is routed through the oxidizer
(valve 3 in position 1)
b) Inject at least 25 ml of sample with the 30 ml size syringe.
c) Immediately after completing the sample injection, switch the sample loop
valve so that the carrier flows in the forward direction so that the sample
loop is swept through the column (valve 1 in position 2)
SCAQMD Method 25.3 March 2000
40
d) Observe the chromatogram and allow the CO2, CH4, ethylene (if present),
and ethane to be eluted from the column. An example chromatogram is
shown in Figure 25.3-9. The period of time during which this takes place
should be approximately eight minutes.
e) After ethane elutes, switch the column valve to backflush mode so that the
carrier flow reverses direction through the column and elutes organics as a
back-flush peak (valve 2 in position 2).
f) Immediately upon switching to backflush mode heat the column oven using
a pre-set temperature profile so that the backflush elutes at a rate so that the
detector responds in its analytical detection range. The temperature profile
should be approximately a follows: Ramp: Increase at a rate of 50 oC/min
for 2 minutes, Final: 150 ± 5 oC for 5 minutes.
In all of the steps, the valving is such that the effluent from the column is
directed to the oxidation reduction and FID detector system. Detector output for
the back-flush peak is sent to the integrator where a response vs. time curve is
plotted and the area under the back-flush peak is integrated. The switching can
be accomplished by manual or automated valving.
Since NMNEOC is a mixture, this back-flush peak may not be symmetrical;
however, the area under a response vs. time curve is proportional to the amount
of carbon present in the sample.
For high CO2 (3% to 15%) and low back-flush (backflush from
SCAQMD Method 25.3 March 2000
41
before the back-flush peak elutes. Measure the CO2 content separately by an
instrument capable of measuring % levels CO2 such as SCAQMD Method 10.1.
4.9.4 Order of Standard, Background, and Sample Injections
Several standards and backgrounds are run with a batch of samples because of
the difficulty in measuring low level concentrations. The following is the order
in which the standards and backgrounds must be run in relation to the samples.
Each step is run through the full fore-flush and backflush of the procedure as
previously described in Section 4.9.3.
a) Inject laboratory air to condition the system and serve as system check by
comparing to historical injections. The system should be able to detect the
background level of 2 - 10 ppm and be consistent with historical levels,
otherwise repeat the laboratory air injection.
b) Inject CO2 free N2 gas to determine “background nitrogen” for NMNEOC
peak (back-flush). The area counts of the back-flush should be within a
historical acceptable area, otherwise repeat the background nitrogen
injection. Save the background nitrogen chromatogram for background
subtraction on the subsequent runs.
c) Inject three level concentrations of TCA standards (recommended: 3, 10 and
100 ppmv) to create a 3-point calibration curve. Accept the calibration curve
if the measured concentrations are within 10% of the check standards.
Otherwise repeat.
d) After calibration, again inject the background nitrogen.
SCAQMD Method 25.3 March 2000
42
e) Inject a 1 ppmv back flush standard. The measured back-flush concentration
must be within ± 20% of the standard, otherwise repeat the background
nitrogen and the 1 ppm backflush standard.
f) Inject the samples in duplicate (each sample is analyzed two times as in
Section 4.9.3). Analyze low concentration samples before high
concentration samples. No more than eight injections (four duplicates) may
be performed in a single batch. Inject the nitrogen background between
batches and at the end of the sample injections.
g) Run two QC standards, choose concentration levels that bracket the sample
concentrations if possible. If the sample has back-flush concentration close
to 1 ppmv, use 1 ppmv as the lower of the QC standards. The measured
back-flush concentration should be within ± 20% of the standard, otherwise
select a new background nitrogen for baseline subtraction as follows: Select
the new background nitrogen chromatogram by injecting nitrogen followed
by the 1 ppmv back flush standard. When the results of the 1 ppm back-
flush standard are within the ± 20% criteria, the chromatogram for the
nitrogen can be used for baseline subtraction. If necessary, divide a large
batch of samples runs in the same day into small batches, and use different
background nitrogen for baseline subtraction as long as the 1 ppmv back
flush standard before and after that particular batch meet the ± 20% criteria.
In standby mode, the column temperature is set to 190 ± 5 oC. The valving is set
so that air is allowed to flow through the oxidation catalyst in the backflush
mode overnight (valve 1 position 2, valve 2 position 2, valve 3 position 2). This
step is essential for the oxidation catalyst to be at full capacity for the next use.
SCAQMD Method 25.3 March 2000
43
The instrument must not be used unless the oxidation catalyst has been
regenerated in this manner.
4.10 TCA Analysis Quality Assurance (QA) Criteria
Pre and post run QC concentrations must be within 10% the standard concentrations.
The following are the required agreement between duplicate analyses of a sample:
Back-flush concentration (ppmv) % difference from mean
1-3 20
4-6 15
7-12 12
13-30 10
31-50 5
If the duplicate analyses do not fall within the required limit, run a third analysis. If after
the third analysis, the mean does not meet the above requirements, review the instrument
calibration and the baseline nitrogen for errors, make necessary changes, then restart from
the background nitrogen step (Step d).
4.11 TCA Calculations
Calculate the concentration of component in the canister using the following equation :
Cc = Cm x DF x (Pf/Pr)
where
Cc = end of sampling canister concentration, ppmv
Cm = average of duplicate measured concentrations from TCA analysis, ppmv
DF = syringe dilution factor if applicable
Pf = canister pressure after pressurization, mm Hg
Pr = before purging canister pressure, mm Hg
SCAQMD Method 25.3 March 2000
44
Calculate the total VOC as equivalent to gaseous carbon using the following equation:
Total VOC (ppmC) = Cw + Cc
where
Cw = VOC from condensate trap water (ppmC)
Cc = VOC from canister(ppmv)
SCAQMD Method 25.3 March 2000
45
PSI
Figure 25.3-5 Inert Gas Purging
SCAQMD Method 25.3 March 2000
46
Figure 25.3-6 Flow Diagram for TCA Analysis on Canisters
SCAQMD Method 25.3 March 2000
47
Figure 25.3-7 Equipment Diagram for TCA Analysis on Canisters
SCAQMD Method 25.3 March 2000
48
Equipment Operation for TCA Analysis on Canisters
Step # Time Valve 1
Position
Valve 2
Position
Valve 3
Position
Description
1 0 1 1 1 Verify Temperatures and Valve Positions
2 0 1 1 1 Inject Sample
3 afterinj.
2 1 1 Switch Carrier Flow Through Sample Loop
4 0-8min.
2 1 1 Observe CO2, CH4, Ethane Elute
5 8 min 2 2 1 Switch to Backflush Mode
6 8-15min
2 2 1 Increase Column Oven Temp by 50oC/min
for 2 min, Hold at 150 oC for 5 min
7 Over-night
2 2 2 Increase Column Oven Temp to 190 oC to
Regenerate Oxidation Catalyst
Figure 25.3-8 Equipment Operation for TCA Analysis on Canisters
SCAQMD Method 25.3 March 2000
49
Figure 25.3-9 Example Chromatogram for TCA Analysis on Canisters
SCAQMD Method 25.3 March 2000
50
METHOD 25.3
DETERMINATION OF LOW CONCENTRATIONNON-METHANE NON-ETHANE ORGANIC COMPOUND EMISSIONS
FROM CLEAN FUELED COMBUSTION SOURCES
Section 5 of 5
5. Engineering Calculations and Reporting
Carry out calculations, retaining at least one extra decimal figure beyond that of the acquired data.
Round off figures after the final calculation.
5.1 Data Quality Checks
The results of the duplicate sampling for NMNEOC must not deviate more than 20% from
the average of the two values in order to meet the precision criteria. The results of the
duplicate sampling for carbon monoxide and carbon dioxide must not deviate more than
20% from the average of the two values in order to meet the leak indicator criteria. Field
observations of occurrences that may cause sample bias may be used to invalidate one of
the duplicate samples in the case that the 20% precision criteria is not satisfied. Individual
results cannot, however, be discarded solely on the basis that the results disagree or that
the results are higher than anticipated. Contamination cannot be used to invalidate the
sample without substantial evidence that contamination occurred. Alternatively the lower
value can be discarded for a worst case evaluation.
5.2 VOC Molecular Weight per Carbon Ratio
In order to convert the lab results as carbon to actual mass emissions as VOC. A
molecular weight per carbon ratio must be either measured, calculated or assumed.
Although a qualitative analytical speciation of the VOC using an approved method is
preferable, it is sometimes not easily accomplished, and other times not feasible due to
SCAQMD Method 25.3 March 2000
51
partitioning of the sample into gaseous and condensable fractions. Other times the ratio
can be calculated based on the VOC formulation of materials consumed in, for example, a
coating or printing operation. In these situations, it is acceptable to use information
provided in Material Data Safety Sheets (MSDS), if considered accurate. The use of
MSDS information is generally, however, not considered as sufficiently accurate for
calculating capture efficiencies. It is acceptable for calculating destruction efficiencies
since the molecular weight per carbon ratio cancels out of the calculation when it is
assumed that the ratio remains constant across the control device. In many cases the ratio
can be considered as represented by a surrogate compound that is representative of the
VOC encountered in the process. In the absence of any of the aforementioned
information, common practice has dictated the use of a default ratio of hexane which is
14.36 lb/lb-mol C. Table 25.3-2 lists several general categories of molecular weight per
carbon ratios which have been deemed as acceptable for the specified applications.
Several specific examples are given below:
a. For Coating and Printing Processes:
Most Preferred: Volatile Carbon Analysis from SCAQMD Protocol for
Determination of VOC Capture Efficiency
Calculation: 12 lb/lb mol x % VOC weighted average / % volatile carbon
weighted average (all percents by weight)
Example: Coating #1 VOC = 50%, % volatile carbon = 40%, usage = 100 lb/hr
Coating #2 VOC = 80%, % volatile carbon = 60%, usage = 10 lb/hr
12 lb/lb-mol x [(50% x 100 lb/hr) + (80% x 10 lb/hr)]MW/C =
[(40% x 100 lb/hr) + (60% x 10 lb/hr)]
= 15.13 lb/lb-molC
SCAQMD Method 25.3 March 2000
52
b. For Coating and Printing Processes: 2nd Choice- MSDS or Formulation Information
MSDS formulation is usually given as weight percent of the total coating/solvent
Calculation: MW/C = Σ(MW x mol frac)/ Σ(carbon# x mol frac)
Example: Coating #1 VOC formulation = 10% benzene, 20% formaldehyde
usage = 10 lb/hr
Coating #2 VOC formulation = 30% butanol, 40% ethylene glycol
monoethyl ether (a.k.a. EGMEE, Cellosolve; 2-Ethoxyethanol),
usage = 100 lb/hr
Benzene Usage = (10% x 10 lb/hr) / 100 = 1 lb/hr
Formaldehyde Usage = (20% x 10 lb/hr) / 100 = 2 lb/hr
Butanol Usage = (30% x 100 lb/hr) / 100 = 30 lb/hr
EGMEE Usage = (40% x 100 lb/hr) / 100 = 40 lb/hr
MWbenzene = 78 lb/lb-mol, C#benzene = 6
MWformaldehyde = 30 lb/lb-mol, C#formaldehyde = 1
MWbutanol = 74 lb/lb-mol, C#butanol = 4
MW EGMEE = 90 lb/lb-mol, C# EGMEE = 4
mol fraci = (usagei / MWi) / Σ(usagei / MWi)
Σ(usagei / MWi) = (1 lb/hrbenzene / 78 lb/lb-molbenzene)
+ (2 lb/hr formaldehyde / 30 lb/lb-molformaldehyde)
+ (30 lb/hr butanol / 74 lb/lb-molbutanol)
+ (40 lb/hr EGMEE / 90 lb/lb-mol EGMEE)
= 0.929 lb-molVOC/hr
mol fracbenzene = (1 lb/hr benzene / 78 lb/lb-molbenzene) / 0.929 = 0.014
mol fracform = (2 lb/hr form / 30 lb/lb-molform) / 0.929 = 0.072
mol fracbutanol = (30 lb/hr butanol / 74 lb/lb-molbutanol) / 0.929 = 0.436
mol fracEGMEE = (40 lb/hr EGMEE / 90 lb/lb-mol EGMEE) / 0.929 = 0.478
(78 lb/lb-mol x 0.014) +(30 lb/lb-mol x 0.072) +(74 lb/lb-mol x 0.436) + (90 lb/lb-mol x 0.478)MW/C =
(6 C x 0.014) +(1 C x 0.072) +(4 C x 0.436) +(4 C x 0.478)
MW/C = 20.6 lb/lb-molC
SCAQMD Method 25.3 March 2000
53
c. If the permit or other emissions limit is specified as a specific compound:
Calculation: MW/C# of specified compound
Example: Permit limit is specified as emissions in units of VOC as Hexane
MW = 86.17 lb/lb-mol
C# = 6.000
MW/C = 86.17 lb/lb-mol / 6.000 C
MW/C = 14.36 lb/lb-mol
d. For Combustion of Only Natural Gas Only:
Calculation: MW/C# of Methane or Hexane. Methane is preferable when
either a worst case emission rate is desired or formaldehyde by-products may be
present due to incomplete combustion. Incomplete combustion may be indicated
by unusually high levels of methane or carbon monoxide.
Example: as Methane.
MW = 16.04 lb/lb-mol
C# = 1.000
MW/C = 16.04 lb/lb-mol / 1.000 C
MW/C = 16.04 lb/lb-mol
e. For fugitive emissions from petroleum processing operations:
Calculation: MW/C# of propane or other compounds if known
Example: as propane
MW = 44.10 lb/lb-mol
C# = 3.000
MW/C = 44.10 lb/lb-mol / 3.000 C
MW/C = 14.70 lb/lb-mol
SCAQMD Method 25.3 March 2000
54
e. For processes that use strictly petroleum distillates:
Calculation: ((14 x C#) +2) / C#
Example: for an average carbon number of 8.
MW/C = ((14 x 8) +2) / 8
MW/C = 14.3 lb/lb-mol
In absence of any information regarding the composition of the VOC, generally the
molecular weight per carbon ratio of hexane is assumed (14.36 lb/lb-molC). As for the
general range of molecular weight to carbon ratios for most VOC mixtures encountered,
formaldehyde (30.03 lb/lb-molC) and methanol (32.04 lb/lb-molC) represent the upper
bounds, while benzene represents the lower bound (13.02 lb/lb-molC). In applications
where the molecular weight per carbon ratio is either difficult to determine as above or in
dispute, a worst case scenario can be used for compliance purposes .
5.3 Bias Correction Factor
During the USEPA’s Office of Air Quality Planning and Standards (OAQPS) evaluation
of this source test method, it was determined that a bias correction factor must be applied
to all results achieved by this method. This correction factor of 1.086 was determined
according to USEPA Method 301 for validating test methods and was based on the results
of the validation testing. The calculation is performed as follows:
Corrected Conc. (ppmC) = Total VOC (ppmC as determined in Section 4.11) x 1.086
SCAQMD Method 25.3 March 2000
55
5.4 VOC Mass Emission Rate Calculation
The individual VOC mass emission rates are determined using the following quantities for
each duct or stack where both a concentration and corresponding flow rate are determined:
C - Average Corrected Concentration of Non-Methane Non-Ethane Organic
Compounds (NMNEOC) from the Method 25.3 sampling pairs reported in
ppmC;
Q - Volumetric flow rate as determined by SCAQMD Methods 1.1 through 4.1 in
dry standard cubic feet per minute;
MW - Molecular Weight per Carbon Ratio as determined in Section 5.2 in lb/lb-mol C;
The VOC mass emissions rate in pounds per hour can then be calculated as follows:
VOC Mass Emission Rate (lb/hr) = 1.583 x 10-7 x MW x C x Q
These calculations can be performed by using the calculation sheet in Figure 1. If multiple
emission points are present, the VOC mass emission rate must be calculated separately as
above and added together for a total VOC mass emission rate.
SCAQMD Method 25.3 March 2000
56
Application Method Calculation MW/C Ratio Typical RangeCoating and
PrintingOperations
% Volatile CarbonAnalysis from
SCAQMDProtocol for
Determination ofVOC Capture
Efficiency
12 lb/lb mol x %VOC weighted
average / %volatile carbon
weighted average
Varies 13-32 lb/lb-mol C
Coating andPrinting
Operations
MSDSInformation
Σ(MW x molfrac)/ Σ(carbon# x
mol frac)
Varies 13-32 lb/lb-mol C
When PermitSpecifies
Compound to beReported as:
SpecifiedCompound
MW/C# Varies 13-32 lb/lb-mol C
Natural Gas/FuelGas Combustion
Assume Hexane MW/C# 14.36 lb/lb-mol C 14.36 lb/lb-mol C
Natural Gas/FuelGas Combustionfor Worst Case or
IncompleteCombustion
Assume Methane(although non-VOC,
sometimes used, accounts
for formaldehyde
formation)
MW/C# 16.04 lb/lb-mol C 16.04 lb/lb-mol C
FugitiveEmissions from
PetroleumProcessingOperations
Assume Propane MW/C# 14.70 lb/lb-mol C 14.70 lb/lb-mol C
Ethanol OnlyProcesses (ethanol
combustion,investment
casting,flexographicprocesses)
Assume Ethanol MW/C# 23.03 lb/lb-mol C 23.03 lb/lb-mol C
Processes ThatUse StrictlyPetroleumDistillates
Average CarbonNumber
((14 x C#) + 2)C#
Varies 14-15 lb/lb-mol C
Processes wherethe Formulation of
the VOC isKnown
VOC Formulation Σ(MW x molfrac)/ Σ(carbon# x
mol frac)
Varies 13-32 lb/lb-mol C
In Absence ofInformation for
Applying Any ofthe Above
Assume Hexane MW/C# 14.36 lb/lb-mol C 14.36 lb/lb-mol C
SCAQMD Method 25.3 March 2000
57
Figure 25.3-10 Molecular Weight per Carbon Ratios
Test No. Date
SOURCE TEST CALCULATIONS
DuctIdentification
Flow Rate(dscfm)
NMNEOC Conc.(ppm)
VOC MassRate (lb/hr)
#1
#2
#3
#4
TOTAL N/A
WHERE:
VOC Mass Rate =
1.583 x 10-7
x (Flow Rate dscfm) x (NMNEOC ppm) (MW per Carbon Ratio lb/lb-mol C)
FIGURE 25.3-11VOC MASS EMISSION RATE CALCULATION
List of Figures 31.Overview and Applicability1.1Principle 41.2Applicability 6
1.4 Limitations and Interferences 82.Sampling Apparatus and Field Equipment Preparation2.1Sampling Apparatus 92.2Sampling Reagents122.3Sampling Equipment Preparation 13
3.Field Procedures3.1Pre-Test Determinations183.2On-site Equipment Assembly183.3Pre-test Leak Check203.4Sampling Operation203.5Reference Point Velocity203.6Post-test Procedures22
4.Laboratory Procedures 4.1Sample Receipt254.2Sample Purge264.3Apparatus and Reagents for TOC Analysis on Traps274.4Preparation of Standards and Reagents for TOC Analysis284.5TOC Analysis on Traps 29
4.6TOC Analysis Quality Assurance (QA) Criteria314.7TOC Calculations324.8Apparatus and Reagents for TCA Analysis on Canisters334.9TCA Analysis on Canisters37
4.10TCA Analysis Quality Assurance (QA) Criteria434.11TCA Calculations43
5.Engineering Calculations and Reporting5.1Data Quality Checks505.2VOC Molecular Weight per Carbon Ratio505.3Bias Correction Factor545.4VOC Mass Emission Rate Calculation55
Figure 25.3-1 Preparation of Sampling Apparatus16Figure 25.3-2 Condensate Trap Detail17Figure 25.3-3 Sampling Apparatus During Sampling23Figure 25.3-4 Field Data Sheet24Figure 25.3-5 Inert Gas Purging45Figure 25.3-6 Flow Diagram for TCA Analysis on Canisters46Figure 25.3-7 Equipment Diagram for TCA Analysis 47Figure 25.3-8 Equipment Operation for TCA Analysis48Figure 25.3-9 Example Chromatogram for TCA Analysis49Figure 25.3-10 Molecular Weight per Carbon Ratios56Figure 25.3-11 VOC Mass Emission Rate Calculation57
1.Overview and Applicability1.1Principle1.2Applicability1.4Limitations and Interferences
2.Sampling Apparatus and Field Equipment Preparation2.1Sampling Apparatus2.2Sampling Reagents2.3Sampling Equipment PreparationFigure 25.3-1 Preparation of Sampling ApparatusFigure 25.3-2 Condensate Trap Detail
3.Field Procedures3.1Pre-test Determinations3.2On-site Equipment Assembly3.3Pretest Leak Check3.4Sampling Operation3.5Reference Point Velocity3.6Post Test ProceduresFigure 25.3-3 Sampling Apparatus During SamplingTest No.DateCompany NameRecorded bySampling LocationMETHOD 25.3 FIELD DATA SHEETPre-Test Leak Check:Post-Test leak CheckGauge Vacuum/in. HgGauge Vacuum:/in. HgLoss in 60 seconds /in. HgLoss in 60 seconds: /in. HgReference Point #Sample Data Reference Point DataObservationsFigure 25.3-4 Field Data SheetMETHOD 25.3
4.Laboratory Procedures4.1Sample Receipt4.2Sample Purge4.3Apparatus and Reagents for TOC Analysis on the Traps4.3.1Shimadzu TOC-5000 Analyzer
\4.3.2Other Apparatus and Reagents for TOC Analysis
4.4Preparation of Standards and Reagents for TOC Analysis4.5TOC Analysis of Traps4.6TOC Analysis Quality Assurance (QA) Criteria4.7TOC Calculations4.8Apparatus and Reagents for TCA Analysis by GC/FID on Canisters4.8.1TCA System
.4.8.2Other Apparatus and Reagents for TCA Analysis
4.9TCA Analysis on the Canisters4.9.1Instrument Parameters and Gas Flow Rates4.9.2Equipment Conditioning4.9.3TCA Procedure4.9.4Order of Standard, Background, and Sample Injections4.10TCA Analysis Quality Assurance (QA) Criteria4.11TCA CalculationsFigure 25.3-5 Inert Gas PurgingFigure 25.3-6 Flow Diagram for TCA Analysis on CanistersFigure 25.3-7 Equipment Diagram for TCA Analysis on CanistersFigure 25.3-8 Equipment Operation for TCA Analysis on CanistersFigure 25.3-9 Example Chromatogram for TCA Analysis on Canisters
5.Engineering Calculations and Reporting5.1Data Quality Checks5.2VOC Molecular Weight per Carbon Ratio5.3Bias Correction Factor5.4VOC Mass Emission Rate Calculation